Clipped conversion from NLP?

Hi,

I’ve recently switched to DSLR as main tool for digitizing my film rolls. I’m deeply satisfied of the results I’m getting, everything looks simpler and better than my previous scans with the Epson Perfection v600.

However, sometimes I’m getting a weird issue: basically, some frames come out with a poor dynamic range or latitude - I would also say kind of bad resolution, despite the rest of the roll being perfectly fine. These weird frames are really hard to manage in LR, namely the sky blows up and pixellates while you’re trying to expose the foreground, and viceversa. Things like that.

Initially, I would think maybe the original picture was underexposed, so the DSLR can’t manage that range. But since I have the chance to look at the same frame that was scanned with the Epson, I can see that it was perfectly scanned, instead! So it looks like the scanner was better in managing the same picture, compared to the DSLR.

So I gave the original RAW files to some friends and they came up with very good results, by using other software than NLP. According to them, NLP has applied way too much contrast and clipped the blue channel. I’m not sure which can cause the issue, but since I’m converting by batch, using same input conditions, something happens during the negative conversion.

GEAR

  1. Negative Lab Pro v.2.3.0 + Lightroom Classic
  2. Canon EOS 60D + Nikon Ai-S Micro NIKKOR 55mm f/2.8

NLP CONVERSION OPTIONS I’M USING

  • WB on black frame - then applied to all the negatives before converting
  • cropping all negatives before converting
  • Frontier
  • Default
  • 5%

Please see below comparative images:

EPSON

Original TIFF file - sorry, for this one I couldn’t find the negative as it came out from the scanner, but the already positive TIFF I made at that time: https://we.tl/t-GJMK4YCuYz

Final JPEG - please note color temperature not edited

DSLR - CANON 60D

Original RAW file: https://we.tl/t-DxYJwSP24L

Working files (snapshots from my LR) - Please note color temperature not edited

(I don’t know if you can see, but it looks like there are patterns on the sky that recall a low resolution image. I don’t know why, since all my frames are digitized with same options (aperture, time, gear) and at the same moment.)

FRIEND’S ATTEMPT USING OTHER CONVERSION THAN NLP
661df7de28df4708943767b96bbe55ed

I hope you may know the issue whether it caused by specific options in NLP. It would be good for me to understand and fix that.

Thank you!

If that is what you see after the NLP conversion, then something is definitely not right! Either a camera profile is missing, or you have some saved default settings that are wacky…

This is what i get just slightly adjusting the white balance (and tweaks to “highs” and “shadows” balancing.

There’s plenty of dynamic range, and no posterization (although the low-res JPEG screenshot I took may be posterized).

If you wanted to go for a brighter, more airy look, you could do something like this…

If you want to edit separate parts of the image separately, you will need to use masks in Lightroom. Negative Lab Pro’s tonal changes will change the tones on the entire image. In this case, the background is very bright at the horizon (I suspect you are shooting into the sun a bit) with the foreground a bit in shadow… the sky is extra tricky here because the upper left of the sky is much darker than the horizon, due to the position of the sun and also probably due to some lens vignetting in your film camera.

For instance, here is an extreme example where I use a gradient mask (with a luminance range filter) on the original RAW after editing in Negative Lab Pro.

Again, that’s an extreme example, and I think the previous two images are better representations of the sceen,

The amount of contrast is completely up to you, and so is the clipping. If you find that NLP is applying too much contrast for your tastes, I’d recommend trying the “Cinematic - Rich” tone profile as your default tone profile. If you find any clipping, you can also use the “WhiteClip and BlackClip” settings to adjust.

Hope that helps!
-Nate

Thanks Nate!

Yeah, there is defintely something wrong, but it only occurs with some pictures when I so the batch conversion. I have now done a single conversion with that specific photo: same NLP input setup, just re-done the WB using the same profile, and this is what I’ve got:

As you can see now everything is normal. So at least I’m glad I don’t have to re-digitilize it or do any trick. Just something wrong with a batch conversion.

Do you think it can be the WB of the first picture of the batch which affect some of the other ones?

You know what? Maybe it might be a missing profile, as you said.
Basically if I do the WB with the Negative Lab v2.3 profile, there is a “Missing profile” written under that. However, the majority of the pictures appear fine, except some of them, as I told you.

If I pick the v2.0 profile, instead, also those weird conversions looks fine.

Ok, yep, you are missing the v2.3 profile.

Are you on Windows, by chance?

In Lightroom, go to “file > import develop presets and profiles” and then navigate to where you downloaded Negative Lab Pro. Add all the files within the “NegativeLab Profiles” folder.

When you do this, you may see that Lightroom will re-render previous conversions. So basically, any conversions you did while the v2.3 profile was missing you will want to either reconvert (open in Negative Lab Pro, go to the “convert” tab and “unconvert”, and then hit “Convert Negatives”). Or, if you’re happy with how some of them were previously, you can just change the profile manually to “Adobe Color” and that should return them to how they were converted without the profile.

Thanks Nate.

I actually had the v2.0 in the meantime, but yes, I reloaded the v2.3 now…something happened it disappeared…